mirror/godot
2
0
Fork 0
mirror of https://github.com/godotengine/godot.git synced 2024-12-27 11:24:59 +08:00
Code Issues Packages Projects Releases Wiki Activity
e9d213e0fd
godot/servers/rendering/rendering_device.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

994 lines
53 KiB
C++
Raw Normal View History

Add static Vulkan loader. Initial Vulkan support for Windows. Initial Vulkan support for macOS.
2019-06-23 00:34:26 +08:00
/*************************************************************************/
/* rendering_device.cpp */
/*************************************************************************/
/* This file is part of: */
/* GODOT ENGINE */
/* https://godotengine.org */
/*************************************************************************/
Update copyright statements to 2022 Happy new year to the wonderful Godot community!
2022-01-04 04:27:34 +08:00
/* Copyright (c) 2007-2022 Juan Linietsky, Ariel Manzur. */
/* Copyright (c) 2014-2022 Godot Engine contributors (cf. AUTHORS.md). */
Add static Vulkan loader. Initial Vulkan support for Windows. Initial Vulkan support for macOS.
2019-06-23 00:34:26 +08:00
/* */
/* Permission is hereby granted, free of charge, to any person obtaining */
/* a copy of this software and associated documentation files (the */
/* "Software"), to deal in the Software without restriction, including */
/* without limitation the rights to use, copy, modify, merge, publish, */
/* distribute, sublicense, and/or sell copies of the Software, and to */
/* permit persons to whom the Software is furnished to do so, subject to */
/* the following conditions: */
/* */
/* The above copyright notice and this permission notice shall be */
/* included in all copies or substantial portions of the Software. */
/* */
/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */
/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */
/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/
/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */
/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */
/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */
/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */
/*************************************************************************/
Initial work on Vulkan: -Added VulkanContext -Added an X11 implementation -Added a rendering device abstraction -added a Vulkan rendering device abstraction -Engine does not work, only shows Godot logo (run it from bin/)
2019-06-08 00:07:57 +08:00
#include "rendering_device.h"
Refactor MethodBind to use variadic templates Removed make_binders and the old style generated binders.
2020-10-15 23:29:59 +08:00
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
#include "rendering_device_binds.h"
Initial work on Vulkan: -Added VulkanContext -Added an X11 implementation -Added a rendering device abstraction -added a Vulkan rendering device abstraction -Engine does not work, only shows Godot logo (run it from bin/)
2019-06-08 00:07:57 +08:00
Replace NULL with nullptr
2020-04-02 07:20:12 +08:00
RenderingDevice *RenderingDevice::singleton = nullptr;
A lot of progress with canvas rendering, still far from working.
2019-06-16 10:45:24 +08:00
RenderingDevice *RenderingDevice::get_singleton() {
return singleton;
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
RenderingDevice::ShaderCompileToSPIRVFunction RenderingDevice::compile_to_spirv_function = nullptr;
Replace NULL with nullptr
2020-04-02 07:20:12 +08:00
RenderingDevice::ShaderCacheFunction RenderingDevice::cache_function = nullptr;
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
RenderingDevice::ShaderSPIRVGetCacheKeyFunction RenderingDevice::get_spirv_cache_key_function = nullptr;
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
void RenderingDevice::shader_set_compile_to_spirv_function(ShaderCompileToSPIRVFunction p_function) {
compile_to_spirv_function = p_function;
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
}
Style: Enforce separation line between function definitions I couldn't find a tool that enforces it, so I went the manual route: ``` find -name "thirdparty" -prune \ -o -name "*.cpp" -o -name "*.h" -o -name "*.m" -o -name "*.mm" \ -o -name "*.glsl" > files perl -0777 -pi -e 's/\n}\n([^#])/\n}\n\n\1/g' $(cat files) misc/scripts/fix_style.sh -c ``` This adds a newline after all `}` on the first column, unless they are followed by `#` (typically `#endif`). This leads to having lots of places with two lines between function/class definitions, but clang-format then fixes it as we enforce max one line of separation. This doesn't fix potential occurrences of function definitions which are indented (e.g. for a helper class defined in a .cpp), but it's better than nothing. Also can't be made to run easily on CI/hooks so we'll have to be careful with new code. Part of #33027.
2020-05-14 20:29:06 +08:00
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
void RenderingDevice::shader_set_spirv_cache_function(ShaderCacheFunction p_function) {
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
cache_function = p_function;
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
void RenderingDevice::shader_set_get_cache_key_function(ShaderSPIRVGetCacheKeyFunction p_function) {
get_spirv_cache_key_function = p_function;
Implement shader caching * Shader compilation is now cached. Subsequent loads take less than a millisecond. * Improved game, editor and project manager startup time. * Editor uses .godot/shader_cache to store shaders. * Game uses user://shader_cache * Project manager uses $config_dir/shader_cache * Options to tweak shader caching in project settings. * Editor path configuration moved from EditorSettings to new class, EditorPaths, so it can be available early on (before shaders are compiled). * Reworked ShaderCompilerRD to ensure deterministic shader code creation (else shader may change and cache will be invalidated). * Added shader compression with SMOLV: https://github.com/aras-p/smol-v
2021-05-25 08:25:11 +08:00
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
Vector<uint8_t> RenderingDevice::shader_compile_spirv_from_source(ShaderStage p_stage, const String &p_source_code, ShaderLanguage p_language, String *r_error, bool p_allow_cache) {
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
if (p_allow_cache && cache_function) {
PoolVector is gone, replaced by Vector Typed `PoolTypeArray` types are now renamed `PackedTypeArray` and are sugar for `Vector<Type>`.
2020-02-18 05:06:54 +08:00
Vector<uint8_t> cache = cache_function(p_stage, p_source_code, p_language);
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
if (cache.size()) {
return cache;
}
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ERR_FAIL_COND_V(!compile_to_spirv_function, Vector<uint8_t>());
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
Adding Variable Rate Shading support to Godot Improve GI renderer and add VRS support Implement render device has_feature and move subgroup settings to limit_get
2022-02-11 19:33:54 +08:00
return compile_to_spirv_function(p_stage, p_source_code, p_language, r_error, this);
Moved the shader source compilation code outside RenderingDevice and Vulkan
2019-07-29 06:58:32 +08:00
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
String RenderingDevice::shader_get_spirv_cache_key() const {
if (get_spirv_cache_key_function) {
Adding Variable Rate Shading support to Godot Improve GI renderer and add VRS support Implement render device has_feature and move subgroup settings to limit_get
2022-02-11 19:33:54 +08:00
return get_spirv_cache_key_function(this);
Implement shader caching * Shader compilation is now cached. Subsequent loads take less than a millisecond. * Improved game, editor and project manager startup time. * Editor uses .godot/shader_cache to store shaders. * Game uses user://shader_cache * Project manager uses $config_dir/shader_cache * Options to tweak shader caching in project settings. * Editor path configuration moved from EditorSettings to new class, EditorPaths, so it can be available early on (before shaders are compiled). * Reworked ShaderCompilerRD to ensure deterministic shader code creation (else shader may change and cache will be invalidated). * Added shader compression with SMOLV: https://github.com/aras-p/smol-v
2021-05-25 08:25:11 +08:00
}
return String();
}
Fixes to mobile renderer * Make sure shaders are named, to aid in debug in case of failure * SceneRenderRD was being wrongly initialized (virtual functions being called when derivative class not initialized). * Fixed some bugs resulting on the above being corrected.
2021-08-17 01:51:29 +08:00
RID RenderingDevice::shader_create_from_spirv(const Vector<ShaderStageSPIRVData> &p_spirv, const String &p_shader_name) {
Vector<uint8_t> bytecode = shader_compile_binary_from_spirv(p_spirv, p_shader_name);
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ERR_FAIL_COND_V(bytecode.size() == 0, RID());
return shader_create_from_bytecode(bytecode);
}
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
RID RenderingDevice::_texture_create(const Ref<RDTextureFormat> &p_format, const Ref<RDTextureView> &p_view, const TypedArray<PackedByteArray> &p_data) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ERR_FAIL_COND_V(p_format.is_null(), RID());
ERR_FAIL_COND_V(p_view.is_null(), RID());
Vector<Vector<uint8_t>> data;
for (int i = 0; i < p_data.size(); i++) {
Vector<uint8_t> byte_slice = p_data[i];
Rename empty() to is_empty()
2020-12-15 20:04:21 +08:00
ERR_FAIL_COND_V(byte_slice.is_empty(), RID());
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
data.push_back(byte_slice);
}
return texture_create(p_format->base, p_view->base, data);
}
RID RenderingDevice::_texture_create_shared(const Ref<RDTextureView> &p_view, RID p_with_texture) {
ERR_FAIL_COND_V(p_view.is_null(), RID());
return texture_create_shared(p_view->base, p_with_texture);
}
Added SSIL post processing effect
2021-08-03 15:07:32 +08:00
RID RenderingDevice::_texture_create_shared_from_slice(const Ref<RDTextureView> &p_view, RID p_with_texture, uint32_t p_layer, uint32_t p_mipmap, uint32_t p_mipmaps, TextureSliceType p_slice_type) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ERR_FAIL_COND_V(p_view.is_null(), RID());
Added SSIL post processing effect
2021-08-03 15:07:32 +08:00
return texture_create_shared_from_slice(p_view->base, p_with_texture, p_layer, p_mipmap, p_mipmaps, p_slice_type);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
RenderingDevice::FramebufferFormatID RenderingDevice::_framebuffer_format_create(const TypedArray<RDAttachmentFormat> &p_attachments, uint32_t p_view_count) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<AttachmentFormat> attachments;
attachments.resize(p_attachments.size());
for (int i = 0; i < p_attachments.size(); i++) {
Ref<RDAttachmentFormat> af = p_attachments[i];
ERR_FAIL_COND_V(af.is_null(), INVALID_FORMAT_ID);
attachments.write[i] = af->base;
}
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
return framebuffer_format_create(attachments, p_view_count);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
RenderingDevice::FramebufferFormatID RenderingDevice::_framebuffer_format_create_multipass(const TypedArray<RDAttachmentFormat> &p_attachments, const TypedArray<RDFramebufferPass> &p_passes, uint32_t p_view_count) {
Vector<AttachmentFormat> attachments;
attachments.resize(p_attachments.size());
for (int i = 0; i < p_attachments.size(); i++) {
Ref<RDAttachmentFormat> af = p_attachments[i];
ERR_FAIL_COND_V(af.is_null(), INVALID_FORMAT_ID);
attachments.write[i] = af->base;
}
Vector<FramebufferPass> passes;
for (int i = 0; i < p_passes.size(); i++) {
Ref<RDFramebufferPass> pass = p_passes[i];
ERR_CONTINUE(pass.is_null());
passes.push_back(pass->base);
}
return framebuffer_format_create_multipass(attachments, passes, p_view_count);
}
RID RenderingDevice::_framebuffer_create(const TypedArray<RID> &p_textures, FramebufferFormatID p_format_check, uint32_t p_view_count) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<RID> textures = Variant(p_textures);
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
return framebuffer_create(textures, p_format_check, p_view_count);
}
RID RenderingDevice::_framebuffer_create_multipass(const TypedArray<RID> &p_textures, const TypedArray<RDFramebufferPass> &p_passes, FramebufferFormatID p_format_check, uint32_t p_view_count) {
Vector<RID> textures = Variant(p_textures);
Vector<FramebufferPass> passes;
for (int i = 0; i < p_passes.size(); i++) {
Ref<RDFramebufferPass> pass = p_passes[i];
ERR_CONTINUE(pass.is_null());
passes.push_back(pass->base);
}
return framebuffer_create_multipass(textures, passes, p_format_check, p_view_count);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
RID RenderingDevice::_sampler_create(const Ref<RDSamplerState> &p_state) {
ERR_FAIL_COND_V(p_state.is_null(), RID());
return sampler_create(p_state->base);
}
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
RenderingDevice::VertexFormatID RenderingDevice::_vertex_format_create(const TypedArray<RDVertexAttribute> &p_vertex_formats) {
Vector<VertexAttribute> descriptions;
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
descriptions.resize(p_vertex_formats.size());
for (int i = 0; i < p_vertex_formats.size(); i++) {
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
Ref<RDVertexAttribute> af = p_vertex_formats[i];
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ERR_FAIL_COND_V(af.is_null(), INVALID_FORMAT_ID);
descriptions.write[i] = af->base;
}
return vertex_format_create(descriptions);
}
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
RID RenderingDevice::_vertex_array_create(uint32_t p_vertex_count, VertexFormatID p_vertex_format, const TypedArray<RID> &p_src_buffers) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<RID> buffers = Variant(p_src_buffers);
return vertex_array_create(p_vertex_count, p_vertex_format, buffers);
}
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
Ref<RDShaderSPIRV> RenderingDevice::_shader_compile_spirv_from_source(const Ref<RDShaderSource> &p_source, bool p_allow_cache) {
ERR_FAIL_COND_V(p_source.is_null(), Ref<RDShaderSPIRV>());
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
Ref<RDShaderSPIRV> bytecode;
Rename `instance()`->`instantiate()` when it's a verb
2021-06-18 06:03:09 +08:00
bytecode.instantiate();
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
for (int i = 0; i < RD::SHADER_STAGE_MAX; i++) {
String error;
ShaderStage stage = ShaderStage(i);
rendering_device: if one compiles with at least on stage without sources, it was always returning an unusable bytecode (because errored)
2021-10-17 03:18:27 +08:00
String source = p_source->get_stage_source(stage);
if (!source.is_empty()) {
Vector<uint8_t> spirv = shader_compile_spirv_from_source(stage, source, p_source->get_language(), &error, p_allow_cache);
bytecode->set_stage_bytecode(stage, spirv);
bytecode->set_stage_compile_error(stage, error);
}
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
return bytecode;
}
Fixes to mobile renderer * Make sure shaders are named, to aid in debug in case of failure * SceneRenderRD was being wrongly initialized (virtual functions being called when derivative class not initialized). * Fixed some bugs resulting on the above being corrected.
2021-08-17 01:51:29 +08:00
Vector<uint8_t> RenderingDevice::_shader_compile_binary_from_spirv(const Ref<RDShaderSPIRV> &p_spirv, const String &p_shader_name) {
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ERR_FAIL_COND_V(p_spirv.is_null(), Vector<uint8_t>());
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
Vector<ShaderStageSPIRVData> stage_data;
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
for (int i = 0; i < RD::SHADER_STAGE_MAX; i++) {
ShaderStage stage = ShaderStage(i);
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ShaderStageSPIRVData sd;
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
sd.shader_stage = stage;
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
String error = p_spirv->get_stage_compile_error(stage);
Replace String comparisons with "", String() to is_empty() Also: - Adds two stress tests to test_string.h - Changes to .empty() on std::strings
2021-12-09 17:42:46 +08:00
ERR_FAIL_COND_V_MSG(!error.is_empty(), Vector<uint8_t>(), "Can't create a shader from an errored bytecode. Check errors in source bytecode.");
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
sd.spir_v = p_spirv->get_stage_bytecode(stage);
Rename empty() to is_empty()
2020-12-15 20:04:21 +08:00
if (sd.spir_v.is_empty()) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
continue;
}
stage_data.push_back(sd);
}
Fixes to mobile renderer * Make sure shaders are named, to aid in debug in case of failure * SceneRenderRD was being wrongly initialized (virtual functions being called when derivative class not initialized). * Fixed some bugs resulting on the above being corrected.
2021-08-17 01:51:29 +08:00
return shader_compile_binary_from_spirv(stage_data, p_shader_name);
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
}
Fixes to mobile renderer * Make sure shaders are named, to aid in debug in case of failure * SceneRenderRD was being wrongly initialized (virtual functions being called when derivative class not initialized). * Fixed some bugs resulting on the above being corrected.
2021-08-17 01:51:29 +08:00
RID RenderingDevice::_shader_create_from_spirv(const Ref<RDShaderSPIRV> &p_spirv, const String &p_shader_name) {
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ERR_FAIL_COND_V(p_spirv.is_null(), RID());
Vector<ShaderStageSPIRVData> stage_data;
for (int i = 0; i < RD::SHADER_STAGE_MAX; i++) {
ShaderStage stage = ShaderStage(i);
ShaderStageSPIRVData sd;
sd.shader_stage = stage;
String error = p_spirv->get_stage_compile_error(stage);
Replace String comparisons with "", String() to is_empty() Also: - Adds two stress tests to test_string.h - Changes to .empty() on std::strings
2021-12-09 17:42:46 +08:00
ERR_FAIL_COND_V_MSG(!error.is_empty(), RID(), "Can't create a shader from an errored bytecode. Check errors in source bytecode.");
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
sd.spir_v = p_spirv->get_stage_bytecode(stage);
if (sd.spir_v.is_empty()) {
continue;
}
stage_data.push_back(sd);
}
return shader_create_from_spirv(stage_data);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Change Array arguments to TypedArray
2022-09-01 01:24:04 +08:00
RID RenderingDevice::_uniform_set_create(const TypedArray<RDUniform> &p_uniforms, RID p_shader, uint32_t p_shader_set) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<Uniform> uniforms;
uniforms.resize(p_uniforms.size());
for (int i = 0; i < p_uniforms.size(); i++) {
Ref<RDUniform> uniform = p_uniforms[i];
ERR_FAIL_COND_V(!uniform.is_valid(), RID());
uniforms.write[i] = uniform->base;
}
return uniform_set_create(uniforms, p_shader, p_shader_set);
}
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
Error RenderingDevice::_buffer_update(RID p_buffer, uint32_t p_offset, uint32_t p_size, const Vector<uint8_t> &p_data, uint32_t p_post_barrier) {
return buffer_update(p_buffer, p_offset, p_size, p_data.ptr(), p_post_barrier);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
static Vector<RenderingDevice::PipelineSpecializationConstant> _get_spec_constants(const TypedArray<RDPipelineSpecializationConstant> &p_constants) {
Vector<RenderingDevice::PipelineSpecializationConstant> ret;
ret.resize(p_constants.size());
for (int i = 0; i < p_constants.size(); i++) {
Ref<RDPipelineSpecializationConstant> c = p_constants[i];
ERR_CONTINUE(c.is_null());
RenderingDevice::PipelineSpecializationConstant &sc = ret.write[i];
Variant value = c->get_value();
switch (value.get_type()) {
case Variant::BOOL: {
sc.type = RD::PIPELINE_SPECIALIZATION_CONSTANT_TYPE_BOOL;
sc.bool_value = value;
} break;
case Variant::INT: {
sc.type = RD::PIPELINE_SPECIALIZATION_CONSTANT_TYPE_INT;
sc.int_value = value;
} break;
case Variant::FLOAT: {
sc.type = RD::PIPELINE_SPECIALIZATION_CONSTANT_TYPE_FLOAT;
sc.float_value = value;
} break;
default: {
}
}
sc.constant_id = c->get_constant_id();
}
return ret;
}
Adding Variable Rate Shading support to Godot Improve GI renderer and add VRS support Implement render device has_feature and move subgroup settings to limit_get
2022-02-11 19:33:54 +08:00
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
RID RenderingDevice::_render_pipeline_create(RID p_shader, FramebufferFormatID p_framebuffer_format, VertexFormatID p_vertex_format, RenderPrimitive p_render_primitive, const Ref<RDPipelineRasterizationState> &p_rasterization_state, const Ref<RDPipelineMultisampleState> &p_multisample_state, const Ref<RDPipelineDepthStencilState> &p_depth_stencil_state, const Ref<RDPipelineColorBlendState> &p_blend_state, int p_dynamic_state_flags, uint32_t p_for_render_pass, const TypedArray<RDPipelineSpecializationConstant> &p_specialization_constants) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
PipelineRasterizationState rasterization_state;
if (p_rasterization_state.is_valid()) {
rasterization_state = p_rasterization_state->base;
}
PipelineMultisampleState multisample_state;
if (p_multisample_state.is_valid()) {
multisample_state = p_multisample_state->base;
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
for (int i = 0; i < p_multisample_state->sample_masks.size(); i++) {
int64_t mask = p_multisample_state->sample_masks[i];
multisample_state.sample_mask.push_back(mask);
}
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
PipelineDepthStencilState depth_stencil_state;
if (p_depth_stencil_state.is_valid()) {
depth_stencil_state = p_depth_stencil_state->base;
}
PipelineColorBlendState color_blend_state;
if (p_blend_state.is_valid()) {
color_blend_state = p_blend_state->base;
Add proper type to most public API uses of Array
2020-04-21 23:16:45 +08:00
for (int i = 0; i < p_blend_state->attachments.size(); i++) {
Ref<RDPipelineColorBlendStateAttachment> attachment = p_blend_state->attachments[i];
if (attachment.is_valid()) {
color_blend_state.attachments.push_back(attachment->base);
}
}
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
return render_pipeline_create(p_shader, p_framebuffer_format, p_vertex_format, p_render_primitive, rasterization_state, multisample_state, depth_stencil_state, color_blend_state, p_dynamic_state_flags, p_for_render_pass, _get_spec_constants(p_specialization_constants));
}
RID RenderingDevice::_compute_pipeline_create(RID p_shader, const TypedArray<RDPipelineSpecializationConstant> &p_specialization_constants = TypedArray<RDPipelineSpecializationConstant>()) {
return compute_pipeline_create(p_shader, _get_spec_constants(p_specialization_constants));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
}
Addition of SDFGI for open world global illumination Move GI to a deferred pass
2020-06-25 21:33:28 +08:00
Vector<int64_t> RenderingDevice::_draw_list_begin_split(RID p_framebuffer, uint32_t p_splits, InitialAction p_initial_color_action, FinalAction p_final_color_action, InitialAction p_initial_depth_action, FinalAction p_final_depth_action, const Vector<Color> &p_clear_color_values, float p_clear_depth, uint32_t p_clear_stencil, const Rect2 &p_region, const TypedArray<RID> &p_storage_textures) {
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<DrawListID> splits;
splits.resize(p_splits);
Addition of SDFGI for open world global illumination Move GI to a deferred pass
2020-06-25 21:33:28 +08:00
Vector<RID> stextures;
for (int i = 0; i < p_storage_textures.size(); i++) {
stextures.push_back(p_storage_textures[i]);
}
draw_list_begin_split(p_framebuffer, p_splits, splits.ptrw(), p_initial_color_action, p_final_color_action, p_initial_depth_action, p_final_depth_action, p_clear_color_values, p_clear_depth, p_clear_stencil, p_region, stextures);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Vector<int64_t> split_ids;
split_ids.resize(splits.size());
for (int i = 0; i < splits.size(); i++) {
split_ids.write[i] = splits[i];
}
return split_ids;
}
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
Vector<int64_t> RenderingDevice::_draw_list_switch_to_next_pass_split(uint32_t p_splits) {
Vector<DrawListID> splits;
splits.resize(p_splits);
Error err = draw_list_switch_to_next_pass_split(p_splits, splits.ptrw());
ERR_FAIL_COND_V(err != OK, Vector<int64_t>());
Vector<int64_t> split_ids;
split_ids.resize(splits.size());
for (int i = 0; i < splits.size(); i++) {
split_ids.write[i] = splits[i];
}
return split_ids;
}
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
void RenderingDevice::_draw_list_set_push_constant(DrawListID p_list, const Vector<uint8_t> &p_data, uint32_t p_data_size) {
ERR_FAIL_COND((uint32_t)p_data.size() > p_data_size);
draw_list_set_push_constant(p_list, p_data.ptr(), p_data_size);
}
void RenderingDevice::_compute_list_set_push_constant(ComputeListID p_list, const Vector<uint8_t> &p_data, uint32_t p_data_size) {
ERR_FAIL_COND((uint32_t)p_data.size() > p_data_size);
compute_list_set_push_constant(p_list, p_data.ptr(), p_data_size);
}
void RenderingDevice::_bind_methods() {
ClassDB::bind_method(D_METHOD("texture_create", "format", "view", "data"), &RenderingDevice::_texture_create, DEFVAL(Array()));
ClassDB::bind_method(D_METHOD("texture_create_shared", "view", "with_texture"), &RenderingDevice::_texture_create_shared);
Added SSIL post processing effect
2021-08-03 15:07:32 +08:00
ClassDB::bind_method(D_METHOD("texture_create_shared_from_slice", "view", "with_texture", "layer", "mipmap", "mipmaps", "slice_type"), &RenderingDevice::_texture_create_shared_from_slice, DEFVAL(1), DEFVAL(TEXTURE_SLICE_2D));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("texture_update", "texture", "layer", "data", "post_barrier"), &RenderingDevice::texture_update, DEFVAL(BARRIER_MASK_ALL));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("texture_get_data", "texture", "layer"), &RenderingDevice::texture_get_data);
ClassDB::bind_method(D_METHOD("texture_is_format_supported_for_usage", "format", "usage_flags"), &RenderingDevice::texture_is_format_supported_for_usage);
ClassDB::bind_method(D_METHOD("texture_is_shared", "texture"), &RenderingDevice::texture_is_shared);
ClassDB::bind_method(D_METHOD("texture_is_valid", "texture"), &RenderingDevice::texture_is_valid);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("texture_copy", "from_texture", "to_texture", "from_pos", "to_pos", "size", "src_mipmap", "dst_mipmap", "src_layer", "dst_layer", "post_barrier"), &RenderingDevice::texture_copy, DEFVAL(BARRIER_MASK_ALL));
ClassDB::bind_method(D_METHOD("texture_clear", "texture", "color", "base_mipmap", "mipmap_count", "base_layer", "layer_count", "post_barrier"), &RenderingDevice::texture_clear, DEFVAL(BARRIER_MASK_ALL));
ClassDB::bind_method(D_METHOD("texture_resolve_multisample", "from_texture", "to_texture", "post_barrier"), &RenderingDevice::texture_resolve_multisample, DEFVAL(BARRIER_MASK_ALL));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_format_create", "attachments", "view_count"), &RenderingDevice::_framebuffer_format_create, DEFVAL(1));
ClassDB::bind_method(D_METHOD("framebuffer_format_create_multipass", "attachments", "passes", "view_count"), &RenderingDevice::_framebuffer_format_create_multipass, DEFVAL(1));
Added GPU based cluster builder Clustering is now GPU based, uses an implementation based on the Activision algorithm.
2021-01-18 00:25:38 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_format_create_empty", "samples"), &RenderingDevice::framebuffer_format_create_empty, DEFVAL(TEXTURE_SAMPLES_1));
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_format_get_texture_samples", "format", "render_pass"), &RenderingDevice::framebuffer_format_get_texture_samples, DEFVAL(0));
ClassDB::bind_method(D_METHOD("framebuffer_create", "textures", "validate_with_format", "view_count"), &RenderingDevice::_framebuffer_create, DEFVAL(INVALID_FORMAT_ID), DEFVAL(1));
ClassDB::bind_method(D_METHOD("framebuffer_create_multipass", "textures", "passes", "validate_with_format", "view_count"), &RenderingDevice::_framebuffer_create_multipass, DEFVAL(INVALID_FORMAT_ID), DEFVAL(1));
Added GPU based cluster builder Clustering is now GPU based, uses an implementation based on the Activision algorithm.
2021-01-18 00:25:38 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_create_empty", "size", "samples", "validate_with_format"), &RenderingDevice::framebuffer_create_empty, DEFVAL(TEXTURE_SAMPLES_1), DEFVAL(INVALID_FORMAT_ID));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_get_format", "framebuffer"), &RenderingDevice::framebuffer_get_format);
Add a Framebuffer cache Adds a FramebufferCache singletion that operates the same way as UniformSetCache. Allows creating framebuffers on the fly (and keep them cached if re-requested) such as: ```C++ RID fb = FramebufferCache::get_singleton()->get_cache(texture1,texture2); ```
2022-08-05 17:59:58 +08:00
ClassDB::bind_method(D_METHOD("framebuffer_is_valid", "framebuffer"), &RenderingDevice::framebuffer_is_valid);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("sampler_create", "state"), &RenderingDevice::_sampler_create);
Reimplement skeletons and blend shapes Uses compute shaders, which only once, on demand, and all in parallel.
2020-12-16 22:07:08 +08:00
ClassDB::bind_method(D_METHOD("vertex_buffer_create", "size_bytes", "data", "use_as_storage"), &RenderingDevice::vertex_buffer_create, DEFVAL(Vector<uint8_t>()), DEFVAL(false));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("vertex_format_create", "vertex_descriptions"), &RenderingDevice::_vertex_format_create);
Expose vertex_array_create
2022-11-04 05:14:55 +08:00
ClassDB::bind_method(D_METHOD("vertex_array_create", "vertex_count", "vertex_format", "src_buffers"), &RenderingDevice::_vertex_array_create);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
doc: Sync classref with current source And fix various bogus bindings following previous PRs.
2021-02-19 20:35:31 +08:00
ClassDB::bind_method(D_METHOD("index_buffer_create", "size_indices", "format", "data", "use_restart_indices"), &RenderingDevice::index_buffer_create, DEFVAL(Vector<uint8_t>()), DEFVAL(false));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("index_array_create", "index_buffer", "index_offset", "index_count"), &RenderingDevice::index_array_create);
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ClassDB::bind_method(D_METHOD("shader_compile_spirv_from_source", "shader_source", "allow_cache"), &RenderingDevice::_shader_compile_spirv_from_source, DEFVAL(true));
Fixes to mobile renderer * Make sure shaders are named, to aid in debug in case of failure * SceneRenderRD was being wrongly initialized (virtual functions being called when derivative class not initialized). * Fixed some bugs resulting on the above being corrected.
2021-08-17 01:51:29 +08:00
ClassDB::bind_method(D_METHOD("shader_compile_binary_from_spirv", "spirv_data", "name"), &RenderingDevice::_shader_compile_binary_from_spirv, DEFVAL(""));
ClassDB::bind_method(D_METHOD("shader_create_from_spirv", "spirv_data", "name"), &RenderingDevice::_shader_create_from_spirv, DEFVAL(""));
Implement Binary Shader Compilation * Added an extra stage before compiling shader, which is generating a binary blob. * On Vulkan, this allows caching the SPIRV reflection information, which is expensive to parse. * On other (future) RenderingDevices, it allows caching converted binary data, such as DXIL or MSL. This PR makes the shader cache include the reflection information, hence editor startup times are significantly improved. I tested this well and it appears to work, and I added a lot of consistency checks, but because it includes writing and reading binary information, rare bugs may pop up, so be aware. There was not much of a choice for storing the reflection information, given shaders can be a lot, take a lot of space and take time to parse.
2021-07-25 22:22:55 +08:00
ClassDB::bind_method(D_METHOD("shader_create_from_bytecode", "binary_data"), &RenderingDevice::shader_create_from_bytecode);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("shader_get_vertex_input_attribute_mask", "shader"), &RenderingDevice::shader_get_vertex_input_attribute_mask);
ClassDB::bind_method(D_METHOD("uniform_buffer_create", "size_bytes", "data"), &RenderingDevice::uniform_buffer_create, DEFVAL(Vector<uint8_t>()));
Environment: Refactor code for readability + more - Makes all boolean setters/getters consistent. - Fixes bug where `glow_hdr_bleed_scale` was not used. - Split CameraEffects to their own source file. - Reorder all Environment method and properties declarations, definitions and bindings to be consistent with each other and with the order of property bindings. - Bind missing enum values added with SDFGI. - Remove unused SDFGI enhance_ssr boolean. - Sync doc changes after SDFGI merge and other misc changes.
2020-07-01 20:18:13 +08:00
ClassDB::bind_method(D_METHOD("storage_buffer_create", "size_bytes", "data", "usage"), &RenderingDevice::storage_buffer_create, DEFVAL(Vector<uint8_t>()), DEFVAL(0));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("texture_buffer_create", "size_bytes", "format", "data"), &RenderingDevice::texture_buffer_create, DEFVAL(Vector<uint8_t>()));
ClassDB::bind_method(D_METHOD("uniform_set_create", "uniforms", "shader", "shader_set"), &RenderingDevice::_uniform_set_create);
ClassDB::bind_method(D_METHOD("uniform_set_is_valid", "uniform_set"), &RenderingDevice::uniform_set_is_valid);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("buffer_update", "buffer", "offset", "size_bytes", "data", "post_barrier"), &RenderingDevice::_buffer_update, DEFVAL(BARRIER_MASK_ALL));
doc: Sync classref with current source
2021-01-28 18:26:37 +08:00
ClassDB::bind_method(D_METHOD("buffer_clear", "buffer", "offset", "size_bytes", "post_barrier"), &RenderingDevice::buffer_clear, DEFVAL(BARRIER_MASK_ALL));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("buffer_get_data", "buffer"), &RenderingDevice::buffer_get_data);
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
ClassDB::bind_method(D_METHOD("render_pipeline_create", "shader", "framebuffer_format", "vertex_format", "primitive", "rasterization_state", "multisample_state", "stencil_state", "color_blend_state", "dynamic_state_flags", "for_render_pass", "specialization_constants"), &RenderingDevice::_render_pipeline_create, DEFVAL(0), DEFVAL(0), DEFVAL(TypedArray<RDPipelineSpecializationConstant>()));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("render_pipeline_is_valid", "render_pipeline"), &RenderingDevice::render_pipeline_is_valid);
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
ClassDB::bind_method(D_METHOD("compute_pipeline_create", "shader", "specialization_constants"), &RenderingDevice::_compute_pipeline_create, DEFVAL(TypedArray<RDPipelineSpecializationConstant>()));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("compute_pipeline_is_valid", "compute_pieline"), &RenderingDevice::compute_pipeline_is_valid);
ClassDB::bind_method(D_METHOD("screen_get_width", "screen"), &RenderingDevice::screen_get_width, DEFVAL(DisplayServer::MAIN_WINDOW_ID));
ClassDB::bind_method(D_METHOD("screen_get_height", "screen"), &RenderingDevice::screen_get_height, DEFVAL(DisplayServer::MAIN_WINDOW_ID));
ClassDB::bind_method(D_METHOD("screen_get_framebuffer_format"), &RenderingDevice::screen_get_framebuffer_format);
ClassDB::bind_method(D_METHOD("draw_list_begin_for_screen", "screen", "clear_color"), &RenderingDevice::draw_list_begin_for_screen, DEFVAL(DisplayServer::MAIN_WINDOW_ID), DEFVAL(Color()));
Fix a few default parameters in bindings They have the wrong type and cause issues with extensions.
2021-07-29 21:42:05 +08:00
ClassDB::bind_method(D_METHOD("draw_list_begin", "framebuffer", "initial_color_action", "final_color_action", "initial_depth_action", "final_depth_action", "clear_color_values", "clear_depth", "clear_stencil", "region", "storage_textures"), &RenderingDevice::draw_list_begin, DEFVAL(Vector<Color>()), DEFVAL(1.0), DEFVAL(0), DEFVAL(Rect2()), DEFVAL(TypedArray<RID>()));
ClassDB::bind_method(D_METHOD("draw_list_begin_split", "framebuffer", "splits", "initial_color_action", "final_color_action", "initial_depth_action", "final_depth_action", "clear_color_values", "clear_depth", "clear_stencil", "region", "storage_textures"), &RenderingDevice::_draw_list_begin_split, DEFVAL(Vector<Color>()), DEFVAL(1.0), DEFVAL(0), DEFVAL(Rect2()), DEFVAL(TypedArray<RID>()));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Add font LCD sub-pixel anti-aliasing support.
2022-08-12 19:03:28 +08:00
ClassDB::bind_method(D_METHOD("draw_list_set_blend_constants", "draw_list", "color"), &RenderingDevice::draw_list_set_blend_constants);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("draw_list_bind_render_pipeline", "draw_list", "render_pipeline"), &RenderingDevice::draw_list_bind_render_pipeline);
ClassDB::bind_method(D_METHOD("draw_list_bind_uniform_set", "draw_list", "uniform_set", "set_index"), &RenderingDevice::draw_list_bind_uniform_set);
ClassDB::bind_method(D_METHOD("draw_list_bind_vertex_array", "draw_list", "vertex_array"), &RenderingDevice::draw_list_bind_vertex_array);
ClassDB::bind_method(D_METHOD("draw_list_bind_index_array", "draw_list", "index_array"), &RenderingDevice::draw_list_bind_index_array);
ClassDB::bind_method(D_METHOD("draw_list_set_push_constant", "draw_list", "buffer", "size_bytes"), &RenderingDevice::_draw_list_set_push_constant);
ClassDB::bind_method(D_METHOD("draw_list_draw", "draw_list", "use_indices", "instances", "procedural_vertex_count"), &RenderingDevice::draw_list_draw, DEFVAL(0));
Fix a few default parameters in bindings They have the wrong type and cause issues with extensions.
2021-07-29 21:42:05 +08:00
ClassDB::bind_method(D_METHOD("draw_list_enable_scissor", "draw_list", "rect"), &RenderingDevice::draw_list_enable_scissor, DEFVAL(Rect2()));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("draw_list_disable_scissor", "draw_list"), &RenderingDevice::draw_list_disable_scissor);
Implement Framebuffer Subpass support * Required for better optimizing mobile platforms * Will be used by the Vulkan mobile renderer.
2021-06-24 21:58:36 +08:00
ClassDB::bind_method(D_METHOD("draw_list_switch_to_next_pass"), &RenderingDevice::draw_list_switch_to_next_pass);
ClassDB::bind_method(D_METHOD("draw_list_switch_to_next_pass_split", "splits"), &RenderingDevice::_draw_list_switch_to_next_pass_split);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("draw_list_end", "post_barrier"), &RenderingDevice::draw_list_end, DEFVAL(BARRIER_MASK_ALL));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Rewrote how barriers work for faster rendering -Added more finegrained control in RenderingDevice API -Optimized barriers (use less ones for thee same) -General optimizations -Shadows render all together unbarriered -GI can render together with shadows. -SDFGI can render together with depth-preoass. -General fixes -Added GPU detection
2021-02-03 03:51:36 +08:00
ClassDB::bind_method(D_METHOD("compute_list_begin", "allow_draw_overlap"), &RenderingDevice::compute_list_begin, DEFVAL(false));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("compute_list_bind_compute_pipeline", "compute_list", "compute_pipeline"), &RenderingDevice::compute_list_bind_compute_pipeline);
ClassDB::bind_method(D_METHOD("compute_list_set_push_constant", "compute_list", "buffer", "size_bytes"), &RenderingDevice::_compute_list_set_push_constant);
ClassDB::bind_method(D_METHOD("compute_list_bind_uniform_set", "compute_list", "uniform_set", "set_index"), &RenderingDevice::compute_list_bind_uniform_set);
ClassDB::bind_method(D_METHOD("compute_list_dispatch", "compute_list", "x_groups", "y_groups", "z_groups"), &RenderingDevice::compute_list_dispatch);
ClassDB::bind_method(D_METHOD("compute_list_add_barrier", "compute_list"), &RenderingDevice::compute_list_add_barrier);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("compute_list_end", "post_barrier"), &RenderingDevice::compute_list_end, DEFVAL(BARRIER_MASK_ALL));
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Rename RD::free binding to free_rid
2021-10-16 03:13:23 +08:00
ClassDB::bind_method(D_METHOD("free_rid", "rid"), &RenderingDevice::free);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("capture_timestamp", "name"), &RenderingDevice::capture_timestamp);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("get_captured_timestamps_count"), &RenderingDevice::get_captured_timestamps_count);
ClassDB::bind_method(D_METHOD("get_captured_timestamps_frame"), &RenderingDevice::get_captured_timestamps_frame);
ClassDB::bind_method(D_METHOD("get_captured_timestamp_gpu_time", "index"), &RenderingDevice::get_captured_timestamp_gpu_time);
ClassDB::bind_method(D_METHOD("get_captured_timestamp_cpu_time", "index"), &RenderingDevice::get_captured_timestamp_cpu_time);
ClassDB::bind_method(D_METHOD("get_captured_timestamp_name", "index"), &RenderingDevice::get_captured_timestamp_name);
ClassDB::bind_method(D_METHOD("limit_get", "limit"), &RenderingDevice::limit_get);
ClassDB::bind_method(D_METHOD("get_frame_delay"), &RenderingDevice::get_frame_delay);
ClassDB::bind_method(D_METHOD("submit"), &RenderingDevice::submit);
ClassDB::bind_method(D_METHOD("sync"), &RenderingDevice::sync);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
ClassDB::bind_method(D_METHOD("barrier", "from", "to"), &RenderingDevice::barrier, DEFVAL(BARRIER_MASK_ALL), DEFVAL(BARRIER_MASK_ALL));
ClassDB::bind_method(D_METHOD("full_barrier"), &RenderingDevice::full_barrier);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
ClassDB::bind_method(D_METHOD("create_local_device"), &RenderingDevice::create_local_device);
doc: Sync classref with current source
2021-01-28 18:26:37 +08:00
ClassDB::bind_method(D_METHOD("set_resource_name", "id", "name"), &RenderingDevice::set_resource_name);
Add named resources and debug labels in RenderDoc
2021-01-24 14:21:54 +08:00
ClassDB::bind_method(D_METHOD("draw_command_begin_label", "name", "color"), &RenderingDevice::draw_command_begin_label);
ClassDB::bind_method(D_METHOD("draw_command_insert_label", "name", "color"), &RenderingDevice::draw_command_insert_label);
ClassDB::bind_method(D_METHOD("draw_command_end_label"), &RenderingDevice::draw_command_end_label);
Rewrote how barriers work for faster rendering -Added more finegrained control in RenderingDevice API -Optimized barriers (use less ones for thee same) -General optimizations -Shadows render all together unbarriered -GI can render together with shadows. -SDFGI can render together with depth-preoass. -General fixes -Added GPU detection
2021-02-03 03:51:36 +08:00
ClassDB::bind_method(D_METHOD("get_device_vendor_name"), &RenderingDevice::get_device_vendor_name);
ClassDB::bind_method(D_METHOD("get_device_name"), &RenderingDevice::get_device_name);
ClassDB::bind_method(D_METHOD("get_device_pipeline_cache_uuid"), &RenderingDevice::get_device_pipeline_cache_uuid);
Fix missing argument names in bindings While at it, tweak some boolean setters to use `p_enabled` for the bool. Also renames `draw_minimap()` to `set_draw_minimap()`.
2021-10-09 05:26:13 +08:00
ClassDB::bind_method(D_METHOD("get_memory_usage", "type"), &RenderingDevice::get_memory_usage);
Fix Render Info * Fixed and redone the process to obtain render information from a viewport * Some stats, such as material changes are too difficult to guess on Vulkan, were removed. * Separated visible and shadow stats, which causes confusion. * Texture, buffer and general video memory can be queried now. * Fixed the performance metrics too.
2021-07-03 07:14:19 +08:00
Expose Vulkan internal values for access from extensions
2021-08-29 10:52:19 +08:00
ClassDB::bind_method(D_METHOD("get_driver_resource", "resource", "rid", "index"), &RenderingDevice::get_driver_resource);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
BIND_CONSTANT(BARRIER_MASK_RASTER);
BIND_CONSTANT(BARRIER_MASK_COMPUTE);
BIND_CONSTANT(BARRIER_MASK_TRANSFER);
BIND_CONSTANT(BARRIER_MASK_ALL);
Rewrote how barriers work for faster rendering -Added more finegrained control in RenderingDevice API -Optimized barriers (use less ones for thee same) -General optimizations -Shadows render all together unbarriered -GI can render together with shadows. -SDFGI can render together with depth-preoass. -General fixes -Added GPU detection
2021-02-03 03:51:36 +08:00
BIND_CONSTANT(BARRIER_MASK_NO_BARRIER);
Reorganize RenderingDevice barriers -Removed sync to draw, now everything syncs to draw by default. -Fixed many validation layer errors. -Added support for VkImageViewUsageCreateInfo to fix validation layer warnings. -Texture, buffer, raster and compute functions now all allow spcifying which barriers will be used.
2021-01-26 08:52:58 +08:00
Add `RenderingServer.get_video_adapter_type()` method This can be used to distinguish between integrated, dedicated, virtual and software-emulated GPUs. This in turn can be used to automatically adjust graphics settings, or warn users about features that may run slowly on their hardware.
2021-12-11 00:01:51 +08:00
BIND_ENUM_CONSTANT(DEVICE_TYPE_OTHER);
BIND_ENUM_CONSTANT(DEVICE_TYPE_INTEGRATED_GPU);
BIND_ENUM_CONSTANT(DEVICE_TYPE_DISCRETE_GPU);
BIND_ENUM_CONSTANT(DEVICE_TYPE_VIRTUAL_GPU);
BIND_ENUM_CONSTANT(DEVICE_TYPE_CPU);
BIND_ENUM_CONSTANT(DEVICE_TYPE_MAX);
Expose Vulkan internal values for access from extensions
2021-08-29 10:52:19 +08:00
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_DEVICE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_PHYSICAL_DEVICE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_INSTANCE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_QUEUE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_QUEUE_FAMILY_INDEX);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_IMAGE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_IMAGE_VIEW);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_IMAGE_NATIVE_TEXTURE_FORMAT);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_SAMPLER);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_DESCRIPTOR_SET);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_BUFFER);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_COMPUTE_PIPELINE);
BIND_ENUM_CONSTANT(DRIVER_RESOURCE_VULKAN_RENDER_PIPELINE);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_ENUM_CONSTANT(DATA_FORMAT_R4G4_UNORM_PACK8);
BIND_ENUM_CONSTANT(DATA_FORMAT_R4G4B4A4_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_B4G4R4A4_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R5G6B5_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_B5G6R5_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R5G5B5A1_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_B5G5R5A1_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_A1R5G5B5_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R8G8B8A8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8A8_SRGB);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_UNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_SNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_USCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_SSCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_UINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_SINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A8B8G8R8_SRGB_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_UNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_SNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_USCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_SSCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_UINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2R10G10B10_SINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_UNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_SNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_USCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_SSCALED_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_UINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_A2B10G10R10_SINT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_SNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_USCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_SSCALED);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R16G16B16A16_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32A32_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32A32_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R32G32B32A32_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64A64_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64A64_SINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_R64G64B64A64_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_B10G11R11_UFLOAT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_E5B9G9R9_UFLOAT_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_D16_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_X8_D24_UNORM_PACK32);
BIND_ENUM_CONSTANT(DATA_FORMAT_D32_SFLOAT);
BIND_ENUM_CONSTANT(DATA_FORMAT_S8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_D16_UNORM_S8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_D24_UNORM_S8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_D32_SFLOAT_S8_UINT);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC1_RGB_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC1_RGB_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC1_RGBA_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC1_RGBA_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC2_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC2_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC3_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC3_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC4_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC4_SNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC5_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC5_SNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC6H_UFLOAT_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC6H_SFLOAT_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC7_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_BC7_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8A1_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8A1_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8A8_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ETC2_R8G8B8A8_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_EAC_R11_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_EAC_R11_SNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_EAC_R11G11_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_EAC_R11G11_SNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_4x4_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_4x4_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_5x4_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_5x4_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_5x5_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_5x5_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_6x5_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_6x5_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_6x6_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_6x6_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x5_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x5_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x6_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x6_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x8_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_8x8_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x5_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x5_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x6_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x6_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x8_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x8_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x10_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_10x10_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_12x10_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_12x10_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_12x12_UNORM_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_ASTC_12x12_SRGB_BLOCK);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8B8G8R8_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B8G8R8G8_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8_B8_R8_3PLANE_420_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8_B8R8_2PLANE_420_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8_B8_R8_3PLANE_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8_B8R8_2PLANE_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G8_B8_R8_3PLANE_444_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_R10X6_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R10X6G10X6_UNORM_2PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R10X6G10X6B10X6A10X6_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6B10X6G10X6R10X6_422_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_B10X6G10X6R10X6G10X6_422_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6_B10X6_R10X6_3PLANE_420_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6_B10X6_R10X6_3PLANE_422_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G10X6_B10X6_R10X6_3PLANE_444_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R12X4_UNORM_PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R12X4G12X4_UNORM_2PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_R12X4G12X4B12X4A12X4_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4B12X4G12X4R12X4_422_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_B12X4G12X4R12X4G12X4_422_UNORM_4PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4_B12X4_R12X4_3PLANE_420_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4_B12X4R12X4_2PLANE_420_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4_B12X4_R12X4_3PLANE_422_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G12X4_B12X4_R12X4_3PLANE_444_UNORM_3PACK16);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16B16G16R16_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_B16G16R16G16_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16_B16_R16_3PLANE_420_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16_B16R16_2PLANE_420_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16_B16_R16_3PLANE_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16_B16R16_2PLANE_422_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_G16_B16_R16_3PLANE_444_UNORM);
BIND_ENUM_CONSTANT(DATA_FORMAT_MAX);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_1D);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_2D);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_3D);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_CUBE);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_1D_ARRAY);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_2D_ARRAY);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_CUBE_ARRAY);
BIND_ENUM_CONSTANT(TEXTURE_TYPE_MAX);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_1);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_2);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_4);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_8);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_16);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_32);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_64);
BIND_ENUM_CONSTANT(TEXTURE_SAMPLES_MAX);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_SAMPLING_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_COLOR_ATTACHMENT_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_DEPTH_STENCIL_ATTACHMENT_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_STORAGE_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_STORAGE_ATOMIC_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_CPU_READ_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_CAN_UPDATE_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_CAN_COPY_FROM_BIT);
BIND_ENUM_CONSTANT(TEXTURE_USAGE_CAN_COPY_TO_BIT);
Use subpasses to do 3D rendering and resolve in mobile renderer
2021-06-26 18:49:25 +08:00
BIND_ENUM_CONSTANT(TEXTURE_USAGE_INPUT_ATTACHMENT_BIT);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_IDENTITY);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_ZERO);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_ONE);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_R);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_G);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_B);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_A);
BIND_ENUM_CONSTANT(TEXTURE_SWIZZLE_MAX);
BIND_ENUM_CONSTANT(TEXTURE_SLICE_2D);
BIND_ENUM_CONSTANT(TEXTURE_SLICE_CUBEMAP);
BIND_ENUM_CONSTANT(TEXTURE_SLICE_3D);
BIND_ENUM_CONSTANT(SAMPLER_FILTER_NEAREST);
BIND_ENUM_CONSTANT(SAMPLER_FILTER_LINEAR);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_REPEAT);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_MIRRORED_REPEAT);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_CLAMP_TO_EDGE);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_CLAMP_TO_BORDER);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_MIRROR_CLAMP_TO_EDGE);
BIND_ENUM_CONSTANT(SAMPLER_REPEAT_MODE_MAX);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_INT_TRANSPARENT_BLACK);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_FLOAT_OPAQUE_BLACK);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_INT_OPAQUE_BLACK);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_FLOAT_OPAQUE_WHITE);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_INT_OPAQUE_WHITE);
BIND_ENUM_CONSTANT(SAMPLER_BORDER_COLOR_MAX);
BIND_ENUM_CONSTANT(VERTEX_FREQUENCY_VERTEX);
BIND_ENUM_CONSTANT(VERTEX_FREQUENCY_INSTANCE);
BIND_ENUM_CONSTANT(INDEX_BUFFER_FORMAT_UINT16);
BIND_ENUM_CONSTANT(INDEX_BUFFER_FORMAT_UINT32);
Addition of SDFGI for open world global illumination Move GI to a deferred pass
2020-06-25 21:33:28 +08:00
BIND_ENUM_CONSTANT(STORAGE_BUFFER_USAGE_DISPATCH_INDIRECT);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_ENUM_CONSTANT(UNIFORM_TYPE_SAMPLER); //for sampling only (sampler GLSL type)
BIND_ENUM_CONSTANT(UNIFORM_TYPE_SAMPLER_WITH_TEXTURE); // for sampling only); but includes a texture); (samplerXX GLSL type)); first a sampler then a texture
BIND_ENUM_CONSTANT(UNIFORM_TYPE_TEXTURE); //only texture); (textureXX GLSL type)
BIND_ENUM_CONSTANT(UNIFORM_TYPE_IMAGE); // storage image (imageXX GLSL type)); for compute mostly
BIND_ENUM_CONSTANT(UNIFORM_TYPE_TEXTURE_BUFFER); // buffer texture (or TBO); textureBuffer type)
BIND_ENUM_CONSTANT(UNIFORM_TYPE_SAMPLER_WITH_TEXTURE_BUFFER); // buffer texture with a sampler(or TBO); samplerBuffer type)
BIND_ENUM_CONSTANT(UNIFORM_TYPE_IMAGE_BUFFER); //texel buffer); (imageBuffer type)); for compute mostly
BIND_ENUM_CONSTANT(UNIFORM_TYPE_UNIFORM_BUFFER); //regular uniform buffer (or UBO).
BIND_ENUM_CONSTANT(UNIFORM_TYPE_STORAGE_BUFFER); //storage buffer ("buffer" qualifier) like UBO); but supports storage); for compute mostly
BIND_ENUM_CONSTANT(UNIFORM_TYPE_INPUT_ATTACHMENT); //used for sub-pass read/write); for mobile mostly
BIND_ENUM_CONSTANT(UNIFORM_TYPE_MAX);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_POINTS);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_LINES);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_LINES_WITH_ADJACENCY);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_LINESTRIPS);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_LINESTRIPS_WITH_ADJACENCY);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TRIANGLES);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TRIANGLES_WITH_ADJACENCY);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TRIANGLE_STRIPS);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TRIANGLE_STRIPS_WITH_AJACENCY);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TRIANGLE_STRIPS_WITH_RESTART_INDEX);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_TESSELATION_PATCH);
BIND_ENUM_CONSTANT(RENDER_PRIMITIVE_MAX);
BIND_ENUM_CONSTANT(POLYGON_CULL_DISABLED);
BIND_ENUM_CONSTANT(POLYGON_CULL_FRONT);
BIND_ENUM_CONSTANT(POLYGON_CULL_BACK);
BIND_ENUM_CONSTANT(POLYGON_FRONT_FACE_CLOCKWISE);
BIND_ENUM_CONSTANT(POLYGON_FRONT_FACE_COUNTER_CLOCKWISE);
BIND_ENUM_CONSTANT(STENCIL_OP_KEEP);
BIND_ENUM_CONSTANT(STENCIL_OP_ZERO);
BIND_ENUM_CONSTANT(STENCIL_OP_REPLACE);
BIND_ENUM_CONSTANT(STENCIL_OP_INCREMENT_AND_CLAMP);
BIND_ENUM_CONSTANT(STENCIL_OP_DECREMENT_AND_CLAMP);
BIND_ENUM_CONSTANT(STENCIL_OP_INVERT);
BIND_ENUM_CONSTANT(STENCIL_OP_INCREMENT_AND_WRAP);
BIND_ENUM_CONSTANT(STENCIL_OP_DECREMENT_AND_WRAP);
BIND_ENUM_CONSTANT(STENCIL_OP_MAX); //not an actual operator); just the amount of operators :D
BIND_ENUM_CONSTANT(COMPARE_OP_NEVER);
BIND_ENUM_CONSTANT(COMPARE_OP_LESS);
BIND_ENUM_CONSTANT(COMPARE_OP_EQUAL);
BIND_ENUM_CONSTANT(COMPARE_OP_LESS_OR_EQUAL);
BIND_ENUM_CONSTANT(COMPARE_OP_GREATER);
BIND_ENUM_CONSTANT(COMPARE_OP_NOT_EQUAL);
BIND_ENUM_CONSTANT(COMPARE_OP_GREATER_OR_EQUAL);
BIND_ENUM_CONSTANT(COMPARE_OP_ALWAYS);
BIND_ENUM_CONSTANT(COMPARE_OP_MAX);
BIND_ENUM_CONSTANT(LOGIC_OP_CLEAR);
BIND_ENUM_CONSTANT(LOGIC_OP_AND);
BIND_ENUM_CONSTANT(LOGIC_OP_AND_REVERSE);
BIND_ENUM_CONSTANT(LOGIC_OP_COPY);
BIND_ENUM_CONSTANT(LOGIC_OP_AND_INVERTED);
BIND_ENUM_CONSTANT(LOGIC_OP_NO_OP);
BIND_ENUM_CONSTANT(LOGIC_OP_XOR);
BIND_ENUM_CONSTANT(LOGIC_OP_OR);
BIND_ENUM_CONSTANT(LOGIC_OP_NOR);
BIND_ENUM_CONSTANT(LOGIC_OP_EQUIVALENT);
BIND_ENUM_CONSTANT(LOGIC_OP_INVERT);
BIND_ENUM_CONSTANT(LOGIC_OP_OR_REVERSE);
BIND_ENUM_CONSTANT(LOGIC_OP_COPY_INVERTED);
BIND_ENUM_CONSTANT(LOGIC_OP_OR_INVERTED);
BIND_ENUM_CONSTANT(LOGIC_OP_NAND);
BIND_ENUM_CONSTANT(LOGIC_OP_SET);
BIND_ENUM_CONSTANT(LOGIC_OP_MAX); //not an actual operator); just the amount of operators :D
BIND_ENUM_CONSTANT(BLEND_FACTOR_ZERO);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE);
BIND_ENUM_CONSTANT(BLEND_FACTOR_SRC_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_SRC_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_DST_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_DST_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_SRC_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_SRC_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_DST_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_DST_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_CONSTANT_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_CONSTANT_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_CONSTANT_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_CONSTANT_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_SRC_ALPHA_SATURATE);
BIND_ENUM_CONSTANT(BLEND_FACTOR_SRC1_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_SRC1_COLOR);
BIND_ENUM_CONSTANT(BLEND_FACTOR_SRC1_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_ONE_MINUS_SRC1_ALPHA);
BIND_ENUM_CONSTANT(BLEND_FACTOR_MAX);
BIND_ENUM_CONSTANT(BLEND_OP_ADD);
BIND_ENUM_CONSTANT(BLEND_OP_SUBTRACT);
BIND_ENUM_CONSTANT(BLEND_OP_REVERSE_SUBTRACT);
BIND_ENUM_CONSTANT(BLEND_OP_MINIMUM);
BIND_ENUM_CONSTANT(BLEND_OP_MAXIMUM);
BIND_ENUM_CONSTANT(BLEND_OP_MAX);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_LINE_WIDTH);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_DEPTH_BIAS);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_BLEND_CONSTANTS);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_DEPTH_BOUNDS);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_STENCIL_COMPARE_MASK);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_STENCIL_WRITE_MASK);
BIND_ENUM_CONSTANT(DYNAMIC_STATE_STENCIL_REFERENCE);
BIND_ENUM_CONSTANT(INITIAL_ACTION_CLEAR); //start rendering and clear the framebuffer (supply params)
Shadow map rendering optimization -All shadow rendering is done with raster now (no compute) -All shadow rendering is done by rendering directly to the shadow atlas -Improved how buffer clearing is done to optimize the above. -Ability to set shadows as 16 bits.
2021-01-25 03:00:20 +08:00
BIND_ENUM_CONSTANT(INITIAL_ACTION_CLEAR_REGION); //start rendering and clear the framebuffer (supply params)
Rewrote how barriers work for faster rendering -Added more finegrained control in RenderingDevice API -Optimized barriers (use less ones for thee same) -General optimizations -Shadows render all together unbarriered -GI can render together with shadows. -SDFGI can render together with depth-preoass. -General fixes -Added GPU detection
2021-02-03 03:51:36 +08:00
BIND_ENUM_CONSTANT(INITIAL_ACTION_CLEAR_REGION_CONTINUE); //continue rendering and clear the framebuffer (supply params)
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_ENUM_CONSTANT(INITIAL_ACTION_KEEP); //start rendering); but keep attached color texture contents (depth will be cleared)
BIND_ENUM_CONSTANT(INITIAL_ACTION_DROP); //start rendering); ignore what is there); just write above it
BIND_ENUM_CONSTANT(INITIAL_ACTION_CONTINUE); //continue rendering (framebuffer must have been left in "continue" state as final action previously)
BIND_ENUM_CONSTANT(INITIAL_ACTION_MAX);
BIND_ENUM_CONSTANT(FINAL_ACTION_READ); //will no longer render to it); allows attached textures to be read again); but depth buffer contents will be dropped (Can't be read from)
BIND_ENUM_CONSTANT(FINAL_ACTION_DISCARD); // discard contents after rendering
BIND_ENUM_CONSTANT(FINAL_ACTION_CONTINUE); //will continue rendering later); attached textures can't be read until re-bound with "finish"
BIND_ENUM_CONSTANT(FINAL_ACTION_MAX);
BIND_ENUM_CONSTANT(SHADER_STAGE_VERTEX);
BIND_ENUM_CONSTANT(SHADER_STAGE_FRAGMENT);
BIND_ENUM_CONSTANT(SHADER_STAGE_TESSELATION_CONTROL);
BIND_ENUM_CONSTANT(SHADER_STAGE_TESSELATION_EVALUATION);
BIND_ENUM_CONSTANT(SHADER_STAGE_COMPUTE);
BIND_ENUM_CONSTANT(SHADER_STAGE_MAX);
BIND_ENUM_CONSTANT(SHADER_STAGE_VERTEX_BIT);
BIND_ENUM_CONSTANT(SHADER_STAGE_FRAGMENT_BIT);
BIND_ENUM_CONSTANT(SHADER_STAGE_TESSELATION_CONTROL_BIT);
BIND_ENUM_CONSTANT(SHADER_STAGE_TESSELATION_EVALUATION_BIT);
BIND_ENUM_CONSTANT(SHADER_STAGE_COMPUTE_BIT);
BIND_ENUM_CONSTANT(SHADER_LANGUAGE_GLSL);
BIND_ENUM_CONSTANT(SHADER_LANGUAGE_HLSL);
Implement Specialization Constants * Added support to our local copy of SpirV Reflect (which does not support it). * Pass them on render or compute pipeline creation. * Not implemented in our shaders yet.
2021-07-10 03:48:28 +08:00
BIND_ENUM_CONSTANT(PIPELINE_SPECIALIZATION_CONSTANT_TYPE_BOOL);
BIND_ENUM_CONSTANT(PIPELINE_SPECIALIZATION_CONSTANT_TYPE_INT);
BIND_ENUM_CONSTANT(PIPELINE_SPECIALIZATION_CONSTANT_TYPE_FLOAT);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_ENUM_CONSTANT(LIMIT_MAX_BOUND_UNIFORM_SETS);
BIND_ENUM_CONSTANT(LIMIT_MAX_FRAMEBUFFER_COLOR_ATTACHMENTS);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURES_PER_UNIFORM_SET);
BIND_ENUM_CONSTANT(LIMIT_MAX_SAMPLERS_PER_UNIFORM_SET);
BIND_ENUM_CONSTANT(LIMIT_MAX_STORAGE_BUFFERS_PER_UNIFORM_SET);
BIND_ENUM_CONSTANT(LIMIT_MAX_STORAGE_IMAGES_PER_UNIFORM_SET);
BIND_ENUM_CONSTANT(LIMIT_MAX_UNIFORM_BUFFERS_PER_UNIFORM_SET);
BIND_ENUM_CONSTANT(LIMIT_MAX_DRAW_INDEXED_INDEX);
BIND_ENUM_CONSTANT(LIMIT_MAX_FRAMEBUFFER_HEIGHT);
BIND_ENUM_CONSTANT(LIMIT_MAX_FRAMEBUFFER_WIDTH);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURE_ARRAY_LAYERS);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURE_SIZE_1D);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURE_SIZE_2D);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURE_SIZE_3D);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURE_SIZE_CUBE);
BIND_ENUM_CONSTANT(LIMIT_MAX_TEXTURES_PER_SHADER_STAGE);
BIND_ENUM_CONSTANT(LIMIT_MAX_SAMPLERS_PER_SHADER_STAGE);
BIND_ENUM_CONSTANT(LIMIT_MAX_STORAGE_BUFFERS_PER_SHADER_STAGE);
BIND_ENUM_CONSTANT(LIMIT_MAX_STORAGE_IMAGES_PER_SHADER_STAGE);
BIND_ENUM_CONSTANT(LIMIT_MAX_UNIFORM_BUFFERS_PER_SHADER_STAGE);
BIND_ENUM_CONSTANT(LIMIT_MAX_PUSH_CONSTANT_SIZE);
BIND_ENUM_CONSTANT(LIMIT_MAX_UNIFORM_BUFFER_SIZE);
BIND_ENUM_CONSTANT(LIMIT_MAX_VERTEX_INPUT_ATTRIBUTE_OFFSET);
BIND_ENUM_CONSTANT(LIMIT_MAX_VERTEX_INPUT_ATTRIBUTES);
BIND_ENUM_CONSTANT(LIMIT_MAX_VERTEX_INPUT_BINDINGS);
BIND_ENUM_CONSTANT(LIMIT_MAX_VERTEX_INPUT_BINDING_STRIDE);
BIND_ENUM_CONSTANT(LIMIT_MIN_UNIFORM_BUFFER_OFFSET_ALIGNMENT);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_SHARED_MEMORY_SIZE);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_COUNT_X);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_COUNT_Y);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_COUNT_Z);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_INVOCATIONS);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_SIZE_X);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_SIZE_Y);
BIND_ENUM_CONSTANT(LIMIT_MAX_COMPUTE_WORKGROUP_SIZE_Z);
Prevent windows from having a size greater than device limit
2022-08-22 06:16:56 +08:00
BIND_ENUM_CONSTANT(LIMIT_MAX_VIEWPORT_DIMENSIONS_X);
BIND_ENUM_CONSTANT(LIMIT_MAX_VIEWPORT_DIMENSIONS_Y);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
Fix Render Info * Fixed and redone the process to obtain render information from a viewport * Some stats, such as material changes are too difficult to guess on Vulkan, were removed. * Separated visible and shadow stats, which causes confusion. * Texture, buffer and general video memory can be queried now. * Fixed the performance metrics too.
2021-07-03 07:14:19 +08:00
BIND_ENUM_CONSTANT(MEMORY_TEXTURES);
BIND_ENUM_CONSTANT(MEMORY_BUFFERS);
BIND_ENUM_CONSTANT(MEMORY_TOTAL);
Exposed RenderingDevice to script API Also added an easier way to load native GLSL shaders. Extras: Had to fix no-cache for subresources in resource loader, it was not properly working, making shaders not properly reload. Note: The precommit hooks are broken because they don't seem to support enums from one class being used in another. Feel free to fix this after merging this PR.
2020-04-20 10:19:21 +08:00
BIND_CONSTANT(INVALID_ID);
BIND_CONSTANT(INVALID_FORMAT_ID);
}
A lot of progress with canvas rendering, still far from working.
2019-06-16 10:45:24 +08:00
RenderingDevice::RenderingDevice() {
Ability to create local RenderingDevice instances.
2020-04-19 07:30:57 +08:00
if (singleton == nullptr) { // there may be more rendering devices later
singleton = this;
}
Initial work on Vulkan: -Added VulkanContext -Added an X11 implementation -Added a rendering device abstraction -added a Vulkan rendering device abstraction -Engine does not work, only shows Godot logo (run it from bin/)
2019-06-08 00:07:57 +08:00
}
Reference in New Issue Copy Permalink